Homing problem after recompile.

boksbox

I use a Kossel Mini delta with arduino Mega 256. I have been using firmware version 0.92.9 for a long time and it's worked fine for me. I decided to upgrade my stepper drivers but the directions needed to be reversed. So I changed the directions using the firmware configuration tool. I uploaded the previous configuration.h file and changed the directions.

I compiled using the latest Arduino IDE and uploaded to the Mega but then it won't home properly. When it hits the end stops, it just stops. When I use Repetier Host, it will Pause and if I click continue it says RequestPause:Homing failed! and
will not continue.

I used the ond configuration.h in hopes of getting everything back but the same problem.

I tried the new 1.0.x firmware and it homes properly but version 1.0.x has other bugs that I'd rather stay with 0.92.9

Help!

Repetier

Did you use INVERT_X_DIR or did you change homing dir? Only the first one is supposed to work correctly.

boksbox

Repetier said:

Did you use INVERT_X_DIR or did you change homing dir? Only the first one is supposed to work correctly.

I checked INVERT*DIR for all towers (was unchecked previously).
I did not change homing dir.

Repetier

Do you have a back move to untrigger end stops for second test? That is a newer addition that you need to untrigger all end stops after first trigger or homing fails. Since you see that message it might only be that then, now the wrong dir idea was excluded already.

boksbox

<a rel="nofollow" target="blank">Repetier</a> said:

Do you have a back move to untrigger end stops for second test? That is a newer addition that you need to untrigger all end stops after first trigger or homing fails. Since you see that message it might only be that then, now the wrong dir idea was excluded already.

On 0.92 - Where can I find back move to untrigger end stops for second test?

I have not changed the previious settings:
ENDSTOP_(X|Y|Z)BACK_MOVE = 5mm
Reduction factor 2nd test = 3mm
ENDSTOP(X|Y|Z)_BACK_ON_HOME = 15MM

boksbox

This is the one I used before. It was working fine. But this time when I recompiled, the homing error occured. Configuration.h > https://drive.google.com/file/d/1ahwOBaQ57jFp025ep6YPzgBw7AO41Vh1/view?usp=sharing

Repetier

It is

#define ENDSTOP_X_BACK_MOVE 5
#define ENDSTOP_Y_BACK_MOVE 5
#define ENDSTOP_Z_BACK_MOVE 5

but 5mm is normally enough I think. You might try 10mm to be safe side especially on mechanical endstops there is a kind where trigger and untrigger differs a bit.

Do you get it on same 0.92 that worked before or are we talking about 1.0.x. Because 0.92 is so old I do not remember if it had already the delta safe homing feature included that seems to trigger here.

If it persists check console and search the error string in sources and you see the exact condition what is wrong. If there are multiple parts with same message make them differ so you see which it is triggering here.

boksbox

The homing error is on 0.92. It has been working before. It would first hit the endstops, go down 5mm and hit it again and will settle to 15mm (back_on_home). I saved the configuration.h before I tried to reverse the steppers.

When the homing error happened after I reversed the steppers, I tried to go back using the saved configuration.h. The same homing error.

I tried 1.0.x and the homing works fine but when I use it, say, try to do baby step Z, the printer stops and will not continue. I have to reset the printer and start over.

I will check console next and see what I can find.

boksbox

Repetier wrote
It is
#define ENDSTOP_X_BACK_MOVE 5
#define ENDSTOP_Y_BACK_MOVE 5
#define ENDSTOP_Z_BACK_MOVE 5
but 5mm is normally enough I think. You might try 10mm to be safe side especially on mechanical endstops there is a kind where trigger and untrigger differs a bit.

Do you get it on same 0.92 that worked before or are we talking about 1.0.x. Because 0.92 is so old I do not remember if it had already the delta safe homing feature included that seems to trigger here.

If it persists check console and search the error string in sources and you see the exact condition what is wrong. If there are multiple parts with same message make them differ so you see which it is triggering here.
c I have tried version 1.0.4 and it homes properly and prints ok. The problem is, when I do filament change or pause the print or even adjust baby-z steps, the printer halts and reset is the only way I could get out of the problem.

I've never had this problem with 0.92 so I'm hoping to keep using it.

I tried again using 10mm Endstop back move. Still stops at first hit and homing fails.

I'm sorry but when you say "check console and search the error string in sources", where do I check it?

I'm using repetier host mac and copied this

10:19:41 AM: Connection opened
< 10:19:42 AM: ústart
< 10:19:42 AM: Transformation matrix: 1.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 1.000000
< 10:19:44 AM: Free RAM:1010
< 10:19:44 AM: Autoretract:0
< 10:19:45 AM: X:0.00 Y:0.00 Z:0.000 E:0.0000
< 10:19:45 AM: SelectExtruder:0
< 10:19:45 AM: FIRMWARE_NAME:Repetier_0.92.9 FIRMWARE_URL:https://github.com/repetier/Repetier-Firmware/ PROTOCOL_VERSION:1.0 MACHINE_TYPE:Delta EXTRUDER_COUNT:1 REPETIER_PROTOCOL:3
< 10:19:45 AM: Printed filament:88.17m Printing time:0 days 13 hours 7 min
< 10:19:45 AM: PrinterMode:FFF
< 10:19:45 AM: DebugLevel:6
< 10:19:45 AM: DebugLevel:6
< 10:20:12 AM: RequestPause:Homing failed!
< 10:20:12 AM: SelectExtruder:0
< 10:20:12 AM: X:0.00 Y:0.00 Z:225.770 E:0.0000

Repetier

With 1.0.4 the ram usage is bigger I think than in 0.92 as it has more functions. Make sure that you have around 1000 free bytes of ram (ide shows free ram after compilation). If there is not enough free ram you overwrite values unplanned and strange things like halt/reset and more will happen. Reduce subsegments per line a bit in case you need more ram.

Console is exactly what you copied. Problem is homing failed might be correct but you need to find where in 0.92 it is issued - I think there are several places to abort homing. You need to find which condition is not met making homing fail. I normally add a print message I would see in console in all possible places.

boksbox

Repetier wrote
With 1.0.4 the ram usage is bigger I think than in 0.92 as it has more functions. Make sure that you have around 1000 free bytes of ram (ide shows free ram after compilation). If there is not enough free ram you overwrite values unplanned and strange things like halt/reset and more will happen. Reduce subsegments per line a bit in case you need more ram.

Console is exactly what you copied. Problem is homing failed might be correct but you need to find where in 0.92 it is issued - I think there are several places to abort homing. You need to find which condition is not met making homing fail. I normally add a print message I would see in console in all possible places.

I reduced subsegments per line as you suggested and it freed up some ram and version 1.0.4 now works without crashing.

Thanks!